Search

Training of the convolution neural network (CNN) is a problem of global optimisation. This study proposed a hybrid modified particle swarm optimisation (MPSO) and conjugate gradient (CG) algorithm for efficient training of CNN. The training involves MPSO–CG to avoid trapping in local minima. Particularly, improvements in the MPSO by introducing a novel approach for control parameters, improved parameters updating criteria, a novel parameter in the velocity update equation, and fusion of the CG allows handling the issues in training CNN. In this study, the authors validate the proposed MPSO algorithm on three benchmark mathematical test functions and also compared with three different variants of the baseline particle swarm optimisation algorithm. Furthermore, the performance of the proposed MPSO–CG is also compared with other training algorithms focusing on the analysis of computational cost, convergence, and accuracy based on a standard problem specific to classification applications on CIFAR-10 dataset and face and skin detection dataset. [ABSTRACT FROM AUTHOR]

The expansion of global optimum methods for electromagnetic design optimization has been successful in the last few years. However, there is no any universal algorithm to be equally successful for all engineering inverse problems. In this regard, inspired from the classical particle swarm optimization (PSO) method and quantum mechanics, this paper presents an improved quantum particle swarm optimizer (MQPSO) by using a tournament selection strategy. Also, a new index, called the torment best (tbest), is incorporated into the QPSO to further enrich its performance. In addition, a new parameter updating strategy is proposed to tradeoff between the exploration and exploitation searches. The feasibility and merit of the proposed approach are verified by the numerical results on mathematic test functions and an electromagnetic inverse problem, namely the TEAM workshop problem 22. [ABSTRACT FROM AUTHOR]

Mutation operator is one of the mechanisms of evolutionary algorithms to guarantee the diversity in the search of an algorithm to help exploring undiscovered search spaces. Thus, in this work, a modified Quantum-inspired Particle Swarm Optimization (QPSO) algorithm for global optimizations of inverse problems is presented. In the proposed algorithm, a new mutation strategy is applied on the personal best particle to improve its global searching ability, also an improved Factor (iF) is incorporated into the position update equation of QPSO to further enhance its convergence speed. In addition, a new parameter updating strategy is proposed to tradeoff between the exploration and exploitation searches. To evaluate its performance, the proposed approach has been applied to a set of well-known mathematical test functions and an engineering inverse problem i.e. TEAM Workshop Problem 22. The experimental results demonstrate the effectiveness and advantage of the proposed method. [ABSTRACT FROM AUTHOR]

The convolution neural network (CNN) has achieved state-of-the-art performance in many computer vision applications e.g., classification, recognition, detection, etc. However, the global optimization of CNN training is still a problem. Fast classification and training play a key role in the development of the CNN. We hypothesize that the smoother and optimized the training of a CNN goes, the more efficient the end result becomes. Therefore, in this paper, we implement a modified resilient backpropagation (MRPROP) algorithm to improve the convergence and efficiency of CNN training. Particularly, a tolerant band is introduced to avoid network overtraining, which is incorporated with the global best concept for weight updating criteria to allow the training algorithm of the CNN to optimize its weights more swiftly and precisely. For comparison, we present and analyze four different training algorithms for CNN along with MRPROP, i.e., resilient backpropagation (RPROP), Levenberg–Marquardt (LM), conjugate gradient (CG), and gradient descent with momentum (GDM). Experimental results showcase the merit of the proposed approach on a public face and skin dataset. [ABSTRACT FROM AUTHOR]